Interactive figures
for 'TSH and FT4 reference interval recommendations: optimization of current methods'
Interactive heatmaps (below) depict the F-scores (diagnostic performance measure of sensitivity and positive predictive value) for various modifications of the non-pregnancy TSH upper limit and the non-pregnancy FT4 lower limit.
Hovering over the figures reveals other diagnostic performance measures for that particular combination of TSH and FT4 reference limit modification.
Please see the original manuscript for a detailed description of the methodology (link will be added after publication).
Performance measures
Sensitivity - true positive rate, true positives among all with a positive test result
Positive Predictive Value (PPV) - true positives among all with a positive test result
Confidence interval (CI) - 95% interval of expected population mean of effect
Prediction interval (PI) - 95% interval of expected individual population effects
I2 statistic - proportion of total variability in effect sizes due to true heterogeneity,
rather than sampling error
Methods used to simulate the current standard 'trimester specific reference intervals':
1) Relative subtraction from upper limit of TSH varying from -5% to -50% in steps
of 5% with subtraction or additions to the lower limit of FT4 varying from -20%
to +20% in steps of 5% (relative subtraction approach),
2) Subtraction from the upper limit of TSH varying from -0.1 to -1.0 mU/L with
subtraction or additions to the lower limit of FT4 varying from -5 to +5 pmol/L
(-0.39 to +0.39 ng/dL; absolute subtraction approach)
3) Using nonpregnancy reference limits with a fixed upper limit for TSH, varying
from 3.0 to 4.5 mU/L, and a fixed lower limit for FT4, varying from 5 to 15
pmol/L (0.39-1.17 ng/dL; fixed limit approach).